Recently, we have investigated the mechanism of diffusion-limited binding of an intrinsically disordered protein (IDP), the N-terminal transactivation domain (TAD) of p53, to one of its binding partners, the nuclear coactivator binding domain (NCBD) of CBP. Diffusion-limited association of IDPs is a counterintuitive phenomenon because it suggests that a disordered protein should fold almost as soon as it encounters with a binding target, which seems very unlikely. We found that a transient complex (TC), which appears during binding, is unexpectedly long lasting (lifetime of several hundred microseconds) due to the stabilization by non-native electrostatic interactions. The long lifetime of TC allows for unstructured TAD to fold without dissociation once it encounters with NCBD, which makes diffusion-limited association possible. Although we have focused on the diffusion-limited association in this study, our experimental observation suggests that the formation of TC by non-native interactions generally occur in IDP binding even for slower binding systems. This work was published in Nature Communications 2018. The next step in this project is to investigate more detailed molecular mechanism of binding. For example, binding pathways are supposed to be heterogeneous as observed in MD simulations. To probe and analyze diverse binding pathways, we use three-color FRET. By attaching the third dye to the binding partner it is possible to detect conformational changes of IDPs and the interaction with binding partners at the same time. Recently, we have developed three-color FRET and fluorescence lifetime analysis of the fast-folding process of a designed protein alpha3D using alternating laser excitation in collaboration with Dr. Irina V. Gopich at LCP (published in J Phys Chem B, 2018). Currently, we are extending this method using continuous-wave laser excitation, which allows for studying protein binding in addition to protein folding at high illumination intensity for high time resolution. In this method, a global analysis of photon trajectories of different color combinations is necessary to determine three FRET efficiencies, which involves determination of a large number of parameters. We implement co-parallelization of CPU-GPU processing for the likelihood calculation, which leads to a significant reduction of the computation time and efficient parameter determination. We have applied this method both for fast folding of alpha3D and fast binding of TAD and NCBD. A manuscript is in preparation. This development of single molecule fluorescence method is also useful for studying oligomerization of proteins that eventually form amyloid fibrils. So far, we have characterized the conformational dynamics of monomers of the two isoforms 40- ad 42-mers, and showed that both proteins are highly disordered (published in Biophysical Journal in 2018). Currently, we are developing single-molecule fluorescence techniques that can probe oligomers and amyloid fibrils both in solution and on a surface.